The present invention relates to a signal processor for multiple gradations for processing an image signal, in order to attain a display with multiple gradations by making up one field of a plurality of subfields weighted with a predetermined luminance and bringing each display cell into a lit or unlit state on a subfield basis, when a display device, such as a PDP panel, EL panel, and some LCD panels, capable of only controlling the lit or unlit state, is used.
In the case of a display device, such as a PDP panel, an EL panel and some LCD panels using a display element which cannot change the intensity of display of each cell arbitrarily, a display with multiple gradations is attained by making up one field of a plurality of subfields weighted with a predetermined luminance and bringing each display cell into a lit or unlit state on a subfield basis.
Conventionally, when a display with multiple gradations is attained by means of the subfield configuration, it is most general that the ratio of luminance among subfields is set to, for example, 1:2:4:8, each term being a power of two. By setting a luminance ratio in such a relation, a display with gradation can be attained most efficiently.
As to a display with multiple gradations using a subfield configuration, there arises the problem that a false contour is caused to occur in the case of a moving image because the lit-state periods are discrete in a field period. In order to solve such a problem, various possible solutions have been proposed heretofore. For example, as to the configuration having the above-mentioned luminance ratio, a method has been proposed in which a subfield with a heavy weight are arranged in the center of a field. With this configuration, however, a moving false contour is caused to occur in an image with relatively small luminance and the image quality is not good enough.
Japanese Unexamined Patent Publication (Kokai) No. 7-271325 has disclosed a superposing method, in which, although the display efficiency is degraded more or less, subfields with the same luminance ratio are provided in one field and plural subfield coding tables for representing the same luminance by a combination of different subfields are prepared and, when a subfield configuration for each pixel is determined, different subfield coding tables are used for each vertical or horizontal display line or each pixel, or for each subfield so that the occurrence of a moving false contour can be prevented.
However, when the superposing method is used, there arises a problem: noise with a checkerboard pattern is produced. Therefore, Japanese Unexamined Patent Publication (Kokai) No. 2002-372948 has proposed a superposing method, in which a moving image and a specific gradation are detected, and the superposing method is used only in a case where that the image is a moving image is detected and that the gradation is likely to produce a moving false contour is detected, and in other cases, one subfield coding table is used.
On the other hand, Japanese Unexamined Patent Publication (Kokai) No. 10-31455 has disclosed a path switching method for preventing the generation of a moving false contour by switching between a sub-path, in which only a gradation whose number of subfields to be lit is increased by one can be displayed and a subfield arrangement in which subfields to be lit are arranged successively is used, and a main path, in which any combination of gradations in such a subfield arrangement can be selected, so that the main path is selected normally for carrying out the process for multiple gradations and the sub-path is selected only in a case where that the image is a moving image is detected and that the gradation is likely to produce a moving false contour is detected. In the path switching method, there exists a difference between the main path and the sub-path in the number of gradations that can be displayed, therefore, the gain adjustment or error diffusion process for an image signal is carried out.
As the superposing method and the path switching method are described in detail in the above-mentioned documents, an explanation will not be given here.